DOI: 10.1118/1.1515762. Lung nodule detection on thoracic computed tomography images: Preliminary evaluation of a computer‐aided diagnosis system

Lung nodule detection on thoracic computed tomography images: Preliminary evaluation of a computer‐aided diagnosis system

10.1118/1.1515762

Crossref journal-article

Wiley

Medical Physics (311)

Abstract

We are developing a computer‐aided diagnosis (CAD) system for lung nodule detection on thoracic helical computed tomography (CT) images. In the first stage of this CAD system, lung regions are identified by a k‐means clustering technique. Each lung slice is classified as belonging to the upper, middle, or the lower part of the lung volume. Within each lung region, structures are segmented again using weighted k‐means clustering. These structures may include true lung nodules and normal structures consisting mainly of blood vessels. Rule‐based classifiers are designed to distinguish nodules and normal structures using 2D and 3D features. After rule‐based classification, linear discriminant analysis (LDA) is used to further reduce the number of false positive (FP) objects. We performed a preliminary study using 1454 CT slices from 34 patients with 63 lung nodules. When only LDA classification was applied to the segmented objects, the sensitivity was 84% (53/63) with 5.48 (7961/1454) FP objects per slice. When rule‐based classification was used before LDA, the free response receiver operating characteristic (FROC) curve improved over the entire sensitivity and specificity ranges of interest. In particular, the FP rate decreased to 1.74 (2530/1454) objects per slice at the same sensitivity. Thus, compared to FP reduction with LDA alone, the inclusion of rule‐based classification lead to an improvement in detection accuracy for the CAD system. These preliminary results demonstrate the feasibility of our approach to lung nodule detection and FP reduction on CT images.

Bibliography

Gurcan, M. N., Sahiner, B., Petrick, N., Chan, H., Kazerooni, E. A., Cascade, P. N., & Hadjiiski, L. (2002). Lung nodule detection on thoracic computed tomography images: Preliminary evaluation of a computerâaided diagnosis system. Medical Physics, 29(11), 2552â2558. Portico.

Authors 7

Metin N. Gurcan (first)
Berkman Sahiner (additional)
Nicholas Petrick (additional)
Heang‐Ping Chan (additional)
Ella A. Kazerooni (additional)
Philip N. Cascade (additional)
Lubomir Hadjiiski (additional)

References 29 Referenced 246

American Cancer Society “American Cancer Society‐Cancer Facts and Figures 2001 ”www.cancer.org(2001).
National Cancer Institute “Surveillance Epidemiology and End Results (SEER) Program Public‐Use CD‐ROM (1973‐1997) ” DCCPS Cancer Surveillance Research Program Cancer Statistics Branch 2000.
10.1378/chest.111.3.754
10.1016/S0140-6736(99)06093-6
10.1016/S0140-6736(97)08229-9
10.1148/radiology.201.3.8939234
10.1148/radiology.199.1.8633131
10.1148/radiology.199.1.8633132
10.1148/radiology.199.1.8633150
10.1148/radiology.212.1.r99jn1461
10.1148/radiol.2203001282
K. Kanazawa Y. Kawata N. Niki H. Satoh H. Ohmatsu R. Kakinuma M. Kaneko N. Moriyama and K. Eguchi “Computer‐aided diagnosis for pulmonary nodules based on helical CT images ” Comput. Med. Imag. Graph. 157–167 (1998). (10.1016/S0895-6111(98)00017-2)
10.1117/12.348530
10.1117/12.348557
10.1117/12.348543
10.1117/12.348611
10.1117/12.387608
10.1117/12.431026
10.1118/1.1387272
10.1148/radiology.218.1.r01ja39267
T. Hara H. Fujita Y. Lee H. Yoshimura and S. Kido Automated lesion detection methods for 2D and 3D chest x‐ray images 27–29 September 1999.
10.1117/12.431145
10.1118/1.597829
W. K. Pratt Digital Imaging Processing 2nd ed. (Wiley New York 1991).
{'key': 'e_1_2_1_26_1', 'first-page': '617', 'article-title': 'Knowledge‐based lung nodule detection from helical CT', 'volume': '205', 'author': 'Erberich S. G.', 'year': '1997', 'journal-title': 'Radiology'} / Radiology / Knowledge‐based lung nodule detection from helical CT by Erberich S. G. (1997)
R. O. Duda and P. E. Hart Pattern Classification and Scene Analysis(Wiley New York 1973).
10.1088/0031-9155/40/5/010
{'key': 'e_1_2_1_29_1', 'first-page': '166', 'article-title': 'A free response approach to the measurement and characterization of radiographic observer performance', 'volume': '4', 'author': 'Bunch P. C.', 'year': '1978', 'journal-title': 'J. Appl. Photogr. Eng.'} / J. Appl. Photogr. Eng. / A free response approach to the measurement and characterization of radiographic observer performance by Bunch P. C. (1978)
10.1109/42.650879

Dates

Type	When
Created	22 years, 9 months ago (Nov. 7, 2002, 3:10 p.m.)
Deposited	1 year, 11 months ago (Sept. 16, 2023, 1:21 a.m.)
Indexed	16 hours, 3 minutes ago (Aug. 24, 2025, 6:56 p.m.)
Issued	22 years, 10 months ago (Oct. 23, 2002)
Published	22 years, 10 months ago (Oct. 23, 2002)
Published Online	22 years, 10 months ago (Oct. 23, 2002)
Published Print	22 years, 9 months ago (Nov. 1, 2002)

Funders 1

U.S. Public Health Service 10.13039/100007197
Region: Americas
gov (National government)
Labels6
1. United States Public Health Service
2. Commissioned Corps of the U.S. Public Health Service
3. USPHS Commissioned Corps
4. U.S. Public Health Service Commissioned Corps
5. USPHS
6. PHS
Awards1
1. CA 93517

BibTeX

@article{Gurcan_2002, title={Lung nodule detection on thoracic computed tomography images: Preliminary evaluation of a computer‐aided diagnosis system}, volume={29}, ISSN={2473-4209}, url={http://dx.doi.org/10.1118/1.1515762}, DOI={10.1118/1.1515762}, number={11}, journal={Medical Physics}, publisher={Wiley}, author={Gurcan, Metin N. and Sahiner, Berkman and Petrick, Nicholas and Chan, Heang‐Ping and Kazerooni, Ella A. and Cascade, Philip N. and Hadjiiski, Lubomir}, year={2002}, month=oct, pages={2552–2558} }

JSON

{
  "indexed": {
    "date-parts": [
      [
        2025,
        8,
        24
      ]
    ],
    "date-time": "2025-08-24T22:56:14Z",
    "timestamp": 1756076174523
  },
  "reference-count": 29,
  "publisher": "Wiley",
  "issue": "11",
  "license": [
    {
      "start": {
        "date-parts": [
          [
            2002,
            10,
            23
          ]
        ],
        "date-time": "2002-10-23T00:00:00Z",
        "timestamp": 1035331200000
      },
      "content-version": "vor",
      "delay-in-days": 0,
      "URL": "http://onlinelibrary.wiley.com/termsAndConditions#vor"
    }
  ],
  "funder": [
    {
      "DOI": "10.13039/100007197",
      "name": "U.S. Public Health Service",
      "doi-asserted-by": "publisher",
      "award": [
        "CA 93517"
      ],
      "id": [
        {
          "id": "10.13039/100007197",
          "id-type": "DOI",
          "asserted-by": "publisher"
        }
      ]
    }
  ],
  "content-domain": {
    "domain": [],
    "crossmark-restriction": false
  },
  "published-print": {
    "date-parts": [
      [
        2002,
        11
      ]
    ]
  },
  "abstract": "<jats:p>We are developing a computer\u2010aided diagnosis (CAD) system for lung nodule detection on thoracic helical computed tomography (CT) images. In the first stage of this CAD system, lung regions are identified by a <jats:italic>k</jats:italic>\u2010means clustering technique. Each lung slice is classified as belonging to the upper, middle, or the lower part of the lung volume. Within each lung region, structures are segmented again using weighted <jats:italic>k</jats:italic>\u2010means clustering. These structures may include true lung nodules and normal structures consisting mainly of blood vessels. Rule\u2010based classifiers are designed to distinguish nodules and normal structures using 2D and 3D features. After rule\u2010based classification, linear discriminant analysis (LDA) is used to further reduce the number of false positive (FP) objects. We performed a preliminary study using 1454 CT slices from 34 patients with 63 lung nodules. When only LDA classification was applied to the segmented objects, the sensitivity was 84% (53/63) with 5.48 (7961/1454) FP objects per slice. When rule\u2010based classification was used before LDA, the free response receiver operating characteristic (FROC) curve improved over the entire sensitivity and specificity ranges of interest. In particular, the FP rate decreased to 1.74 (2530/1454) objects per slice at the same sensitivity. Thus, compared to FP reduction with LDA alone, the inclusion of rule\u2010based classification lead to an improvement in detection accuracy for the CAD system. These preliminary results demonstrate the feasibility of our approach to lung nodule detection and FP reduction on CT images.</jats:p>",
  "DOI": "10.1118/1.1515762",
  "type": "journal-article",
  "created": {
    "date-parts": [
      [
        2002,
        11,
        7
      ]
    ],
    "date-time": "2002-11-07T20:10:45Z",
    "timestamp": 1036699845000
  },
  "page": "2552-2558",
  "source": "Crossref",
  "is-referenced-by-count": 246,
  "title": "Lung nodule detection on thoracic computed tomography images: Preliminary evaluation of a computer\u2010aided diagnosis system",
  "prefix": "10.1002",
  "volume": "29",
  "author": [
    {
      "given": "Metin N.",
      "family": "Gurcan",
      "sequence": "first",
      "affiliation": []
    },
    {
      "given": "Berkman",
      "family": "Sahiner",
      "sequence": "additional",
      "affiliation": []
    },
    {
      "given": "Nicholas",
      "family": "Petrick",
      "sequence": "additional",
      "affiliation": []
    },
    {
      "given": "Heang\u2010Ping",
      "family": "Chan",
      "sequence": "additional",
      "affiliation": []
    },
    {
      "given": "Ella A.",
      "family": "Kazerooni",
      "sequence": "additional",
      "affiliation": []
    },
    {
      "given": "Philip N.",
      "family": "Cascade",
      "sequence": "additional",
      "affiliation": []
    },
    {
      "given": "Lubomir",
      "family": "Hadjiiski",
      "sequence": "additional",
      "affiliation": []
    }
  ],
  "member": "311",
  "published-online": {
    "date-parts": [
      [
        2002,
        10,
        23
      ]
    ]
  },
  "reference": [
    {
      "key": "e_1_2_1_2_1",
      "unstructured": "American Cancer Society \u201cAmerican Cancer Society\u2010Cancer Facts and Figures 2001 \u201dwww.cancer.org(2001)."
    },
    {
      "key": "e_1_2_1_3_1",
      "unstructured": "National Cancer Institute \u201cSurveillance Epidemiology and End Results (SEER) Program Public\u2010Use CD\u2010ROM (1973\u20101997) \u201d DCCPS Cancer Surveillance Research Program Cancer Statistics Branch 2000."
    },
    {
      "key": "e_1_2_1_4_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1378/chest.111.3.754"
    },
    {
      "key": "e_1_2_1_5_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1016/S0140-6736(99)06093-6"
    },
    {
      "key": "e_1_2_1_6_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1016/S0140-6736(97)08229-9"
    },
    {
      "key": "e_1_2_1_7_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1148/radiology.201.3.8939234"
    },
    {
      "key": "e_1_2_1_8_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1148/radiology.199.1.8633131"
    },
    {
      "key": "e_1_2_1_9_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1148/radiology.199.1.8633132"
    },
    {
      "key": "e_1_2_1_10_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1148/radiology.199.1.8633150"
    },
    {
      "key": "e_1_2_1_11_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1148/radiology.212.1.r99jn1461"
    },
    {
      "key": "e_1_2_1_12_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1148/radiol.2203001282"
    },
    {
      "key": "e_1_2_1_13_1",
      "doi-asserted-by": "crossref",
      "unstructured": "K. Kanazawa Y. Kawata N. Niki H. Satoh H. Ohmatsu R. Kakinuma M. Kaneko N. Moriyama and K. Eguchi \u201cComputer\u2010aided diagnosis for pulmonary nodules based on helical CT images \u201d Comput. Med. Imag. Graph. 157\u2013167 (1998).",
      "DOI": "10.1016/S0895-6111(98)00017-2"
    },
    {
      "key": "e_1_2_1_14_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1117/12.348530"
    },
    {
      "key": "e_1_2_1_15_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1117/12.348557"
    },
    {
      "key": "e_1_2_1_16_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1117/12.348543"
    },
    {
      "key": "e_1_2_1_17_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1117/12.348611"
    },
    {
      "key": "e_1_2_1_18_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1117/12.387608"
    },
    {
      "key": "e_1_2_1_19_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1117/12.431026"
    },
    {
      "key": "e_1_2_1_20_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1118/1.1387272"
    },
    {
      "key": "e_1_2_1_21_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1148/radiology.218.1.r01ja39267"
    },
    {
      "key": "e_1_2_1_22_1",
      "unstructured": "T. Hara H. Fujita Y. Lee H. Yoshimura and S. Kido Automated lesion detection methods for 2D and 3D chest x\u2010ray images 27\u201329 September 1999."
    },
    {
      "key": "e_1_2_1_23_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1117/12.431145"
    },
    {
      "key": "e_1_2_1_24_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1118/1.597829"
    },
    {
      "key": "e_1_2_1_25_1",
      "unstructured": "W. K. Pratt Digital Imaging Processing 2nd ed. (Wiley New York 1991)."
    },
    {
      "key": "e_1_2_1_26_1",
      "first-page": "617",
      "article-title": "Knowledge\u2010based lung nodule detection from helical CT",
      "volume": "205",
      "author": "Erberich S. G.",
      "year": "1997",
      "journal-title": "Radiology"
    },
    {
      "key": "e_1_2_1_27_1",
      "unstructured": "R. O. Duda and P. E. Hart Pattern Classification and Scene Analysis(Wiley New York 1973)."
    },
    {
      "key": "e_1_2_1_28_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1088/0031-9155/40/5/010"
    },
    {
      "key": "e_1_2_1_29_1",
      "first-page": "166",
      "article-title": "A free response approach to the measurement and characterization of radiographic observer performance",
      "volume": "4",
      "author": "Bunch P. C.",
      "year": "1978",
      "journal-title": "J. Appl. Photogr. Eng."
    },
    {
      "key": "e_1_2_1_30_1",
      "doi-asserted-by": "publisher",
      "DOI": "10.1109/42.650879"
    }
  ],
  "container-title": "Medical Physics",
  "original-title": [],
  "language": "en",
  "link": [
    {
      "URL": "https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1118%2F1.1515762",
      "content-type": "application/pdf",
      "content-version": "vor",
      "intended-application": "text-mining"
    },
    {
      "URL": "https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1118%2F1.1515762",
      "content-type": "unspecified",
      "content-version": "vor",
      "intended-application": "text-mining"
    },
    {
      "URL": "https://aapm.onlinelibrary.wiley.com/doi/pdf/10.1118/1.1515762",
      "content-type": "unspecified",
      "content-version": "vor",
      "intended-application": "similarity-checking"
    }
  ],
  "deposited": {
    "date-parts": [
      [
        2023,
        9,
        16
      ]
    ],
    "date-time": "2023-09-16T05:21:14Z",
    "timestamp": 1694841674000
  },
  "score": 1,
  "resource": {
    "primary": {
      "URL": "https://aapm.onlinelibrary.wiley.com/doi/10.1118/1.1515762"
    }
  },
  "subtitle": [],
  "short-title": [],
  "issued": {
    "date-parts": [
      [
        2002,
        10,
        23
      ]
    ]
  },
  "references-count": 29,
  "journal-issue": {
    "issue": "11",
    "published-print": {
      "date-parts": [
        [
          2002,
          11
        ]
      ]
    }
  },
  "alternative-id": [
    "10.1118/1.1515762"
  ],
  "URL": "http://dx.doi.org/10.1118/1.1515762",
  "relation": {},
  "ISSN": [
    "0094-2405",
    "2473-4209"
  ],
  "subject": [],
  "container-title-short": "Medical Physics",
  "published": {
    "date-parts": [
      [
        2002,
        10,
        23
      ]
    ]
  }
}